Peer group and distance: when widening university participation is better

We analyze the welfare effect of allowing a new university in a local area where another university is already operating. We use a two-city model in which individuals, whose education depends on the average peer ability (peer group effect), can sort across cities by facing a mobility cost. Com- paring monopoly with a two-university system we find that introducing the second university is always welfare improving. We obtain a symmetric Nash equilibrium for every mobility costs and asymmetric Nash equilibria only for sufficiently low mobility costs. In particular, in the symmetric scenario both universities have the same peer groups (lower than the peer group under monopoly) and the same number of students. The asymmetric scenario instead is such that the "top" ("bottom") university has a peer group higher (lower) than the monopolistic one. Moreover, we find that the symmetric scenario always induces the highest welfare. After checking for equilibrium refinements we find that asymmetric equilibria are never strong Nash whereas the symmetric equilibrium is strong Nash only for sufficiently high mobility costs.